Thursday, September 23, 2010

Tropical Depression #15

Tropical depression #15 has developed this afternoon in the south-central Caribbean as shown on the visible satellite image on the left. The depression has become more organized as thunderstorm activity continues to increase near the center of circulation. The future of the depression in the long-term is quite unknown at this time. The system is forecast by most of the computer models to continue westward, possibly clipping the coast of Nicaragua or Honduras before heading toward the Yucatan peninsula. At this time is when the forecast gets muddy. Most of the models indicate that the cyclone may reach a region of weak steering near the Yucatan. Quite a bit hinges on if the system makes landfall on the peninsula or remains offshore. IF the system remains offshore it has a greater potential to be a significant storm later down the road. The official forecast from the National Hurricane Center does call for the depression to become a hurricane prior to reaching the Yucatan peninsula. Right now most of the computer representations of the atmosphere are struggling to figure out where this storm may go after reaching the Yucatan, due in part to an upper-level low which may get stuck over the southeastern United States. There is some that suggest that the low may be able to turn the cyclone northward into the Gulf of Mexico. It is too early to tell right now but we will monitor this storm closely as it may represent the highest threat to the US since Hurricane Earl several weeks ago.

Wednesday, September 15, 2010

2010 Atlantic Hurricane Season... So far

I'd like to take a blog to discuss a little about the 2010 Atlantic hurricane season to date. Prior to the start of the season, the majority of the agencies, universities, and private groups that provide seasonal forecasts suggested that the 2010 season would see above normal activity. This was not a surprise given that ocean temperatures across the Atlantic basin were and still are significantly above normal and we were entering a La Nina phase in the equatorial Pacific which typically enhances tropical cyclone activity in the Atlantic basin. Although seasonal forecasts from NOAA and Colorado State University have shown some reasonable skill at predicting overall activity, currently there is no skill in determining if a season will feature more landfalls along the US coastline or not.
2010 got off to a quick start with Hurricane Alex, shown making landfall in the radar image to the left. This storm made landfall only a 130 miles south of the Texas/Mexico border as an intensifying category 2 hurricane at approximately 02 UTC on the 1st of July. Alex featured a very low central pressure of 946 mb, which is often supportive of a stronger hurricane, but the wind field had not responded yet and had the system had another 12-24 hours over water it would have likely been a major hurricane. In any event, Alex was the second strongest June hurricane on record. Fortunately for the residents along the lower Texas coast a strong ridge of high pressure over the Gulf of Mexico kept Alex from making landfall further north.

Despite the quick start, the season slowed considerably with short lived storms: Bonnie and Colin. Although the tropical wave activity continued to roll off the African continent, waves had to deal with a variety of inhibiting factors. The most dominant was a very dry eastern Atlantic due in part to a significant amount of African dust, known as the Saharan Air Layer (SAL). Several recent scientific papers have focused on the SAL as both an inhibiting factor and a feature which can aid in tropical development. In this case, the SAL appeared to inhibit development by providing a very dry and stable airmass across much of the main development region. Coupled with the SAL is an easterly jet feature which is often present in the middle layers of the troposphere across the eastern Atlantic. This can induce wind shear on a developing tropical cyclone, thus not allowing the system to organize. Also the presence of a parade of upper-level low pressure areas across much of the central Atlantic increased the wind shear levels as well. All of this contributed to a relative lull in the tropical activity from July through mid August.

Activity quickly picked up as Hurricane Danielle developed. I have stated this before but it seem as if the atmosphere switched the tropical cyclone genesis switch into the "on" position. What would ensue was a parade of tropical cyclones most of which developed from tropical waves emerging off the western coast of Africa. These systems are often referred to as Cape Verde's due to the development region's proximity to the Cape Verde Islands. Many of the major hurricanes in our historical record were of the Cape Verde variety. Danielle would go on to become the first "major" hurricane (cat 3 or higher) of the 2010 season but recurve harmlessly out into the north Atlantic. Danielle was followed by Earl (image below) which made a much closer approach to the US before finally making landfall as a tropical storm in Nova Scotia. Earl, like Danielle, intensified to a major hurricane just after passing near Anegada in the northern Leeward Islands. Earl would eventually pass approximately 100 miles east of Cape Hatteras and the outer banks of North Carolina. Up next were short-lived tropical storms Fiona and Gaston.

Up next was Hermine, a storm that rapidly intensified from a tropical depression to a strong tropical storm in 21 hours in the Bay of Campeche. Hermine came right up to the brink of being classified a hurricane at landfall in Mexico, well south of the Texas/Mexico border. Hermine however reminded us of the threats that even tropical storms can pose as the system proceeded northward after landfall into central Texas. Hermine dumped nearly 10 inches of rainfall across the Austin, TX metro area and rainband convection spawned several tornadoes in the DFW metroplex. Tornadoes associated with tropical systems are typically found in the rainband regions and can occur well after landfall. In fact, one of our graduate students and TTUHRT blogger Scott Gunter is currently analyzing data collected from the SMART-radars during the landfall of Hurricane Frances (2004). His dataset focuses on two individual cells within an outer rainband of Frances that contained rotational signatures.

Moving on in the 2010 Atlantic season brings us to Hurricane Igor, Hurricane Julia, and Tropical Storm Karl which were all active at the time this was written. Igor (image to the right) and Julia both developed from vigorous tropical waves almost as soon as they exited the African coast. Both also went through significant phases of rapid intensification. Whereas Karl developed in the western Caribbean, making landfall as a Tropical Storm near the Belize/Mexico border. Igor and Julia are both forecast to turn northward into the open Atlantic and will likely not threaten the east coast of the United States. Although Igor may be a potential threat to Bermuda. At the time this blog was being written, both Igor and Julia were category 4 hurricanes. The last time two category 4 hurricanes were active in the Atlantic basin was in 1926.

Looking back, most of our Cape Verde systems over the past month have fortunately followed a similar track and recurved out into the open Atlantic and have not been significant threats to the United States. Why is this? well hurricanes are steered by the flow of air over a large depth of the troposphere. They typically move around large and deep areas of high pressure (ridge). Well typically during the peak of the hurricane season a well established ridge is present over the central Atlantic, known as the Bermuda-Azores High. Often tropical systems follow the southern extent of the ridge westward as this is the "path of least resistance". If the ridge does not extent far enough westward the system will simply rotate around the ridge and move more northward. This has been the case this season, as a series of upper-level low pressure areas or troughs have passed through the flow and weaken or erode the western side of the Bermuda High, thus allowing tropical systems to follow the "easy" path northward around the ridge. In 2004 and 2005 this was a different story as the ridge extended far enough westward to simply push systems toward the continental United States. Although this is a very simplistic representation of the steering regimes of tropical cyclones it gives a general idea how Cape Verde storms often progress. Predicting this pattern prior to the season is quite difficult, thus estimating the risk to areas of coastline before the start of a hurricane season is a tremendous forecasting challenge. As far as activity for this season, despite the slow start it is well on its way to being a very active year. So far this season has featured 11 named storms, 5 hurricanes, and 4 major hurricanes (Danielle, Earl, Igor, and Julia), an average season has 10-6-2. So with about 45% of the season remaining it looks like the seasonal forecasts of an active year will verify.

In the next blog, I'd like to highlight a little about the ongoing research within our group here at Texas Tech.

Friday, September 10, 2010

The Story Behind the Texas Tech Hurricane Research Team

We'd like to start the inaugural blog with the story how TTUHRT came to be. Most people would not directly associate Texas Tech University with hurricane research since we are at the closest about 9 hours from the coast and tucked out here in West Texas. Texas Tech however has a rich history in studying severe wind storms. The Wind Science and Engineering Research Center came about following the 1970 Lubbock tornado and has become a leader in wind hazard research. Initial research efforts focused on conducting damage surveys in order to understand the types of building failures that occur in severe windstorms, which include tornadoes, hurricanes, microburst events, etc etc. Research has since moved forward in an effort to answer the question "Is Wind Wind?", meaning are tornadic winds similar to hurricane winds or to microbursts or just a synoptic wind event that we see all the time out here in West Texas. The Wind Science and Engineering Research Center has always emphasized making full-scale measurements of the atmosphere and its effects on man's built environment.

This emphasis on full-scale measurements and understanding the structure of severe winds led to the development of our existing Hurricanes at Landfall Project. In the late 90's, Dr. John Schroeder (then conducting his doctoral dissertation research) went searching for high-resolution data from the eyewall of landfalling hurricanes and was unsuccessful. This is due to typical failure of conventional weather stations used by the National Weather Service (NWS) to fail in sustained wind speeds of only 50 mph. Most believe that the stations are simply destroyed. Well in most cases this is not the truth at all. Unfortunately our conventional network of observing stations is powered by our standard electrical grid with no backup power supply. Also stations do not have an internal data logger and must rely on data transmission back to the NWS. So when the power goes out in an area, so do the observations. Question why 50 mph? these are typical sustained winds that begin to produce damage, such as downed trees which often fall on power lines. A recent study from the Florida Coastal Monitoring Program found an 80% failure rate of conventional automated surface observing stations (ASOS) in sustained winds of 50 mph.

So the question was asked, if there is no data why can't we just go and collect it ourselves? That is exactly what scientists here at TTU set out to do. In 1998 Texas Tech developed its first ruggedized instrumented tower which could be deployed in the path of a landfalling hurricane. The experiment was called the Wind Engineering Mobile Instrument Tower Experiment or WEMITE for short. Dr. Schroeder designed WEMITE #1 (image to the left) to be self-sustaining with its own power supply and be rugged enough to survive the severe winds found in hurricane eyewalls. WEMITE #1 was a telescoping lattice 10 m tower mounted on a trailer platform. The tower was anchored by 6 guy wires using modified mobile home anchors as well as outriggers attached to the trailer which were also attached to the ground through the same type of anchor. The system was powered by a bank of several marine batteries within a debris-proof enclosure. A small wind generator was also added to allow for continuous charging of the battery bank. An internal laptop computer handled the acquisition and storage of data. Multiple wind sensors (anemometers) were place on the tower as well as temperature/relative humidity and barometric pressure sensors. WEMITE #1 was completed in time for the 1998 Atlantic Hurricane season and Texas Tech became the first institution to deploy an instrumented meterological tower in the path of a landfalling hurricane when Hurricane Bonnie made landfall in North Carolina. WEMITE #1 would become the work horse of the fleet making 20 deployments between 1998 and 2005.

In 1999 WEMITE #2 was added. This platform was also a trailer mounted ruggedized tower. Three more single mast non-ruggedized towers were added to the fleet in 2002 bringing TTUHRT's capacity to 5 platforms. During this time the Florida Coastal Monitoring Program (FCMP) was developed by Dr. Tim Reinhold. This was a collaboration between the University of Florida, Clemson, and Florida International to develop and deploy ruggedized towers in the path of landfalling tropical cyclones. FCMP has been one of our closest collaborators over the years and we look forward to continued scientific study between our groups. TTUHRT also joined the SMART-Radar coallition with Oklahoma University, Texas A&M, and the National Severe Storms Laboratory. The two SMART-Radars are C-band mobile Doppler radar systems and provide higher resolution data than conventional NWS Doppler radars (WSR-88D). TTUHRT deployed both radars for Hurricanes Lili (2002), Hurricane Isabel (2003), and Hurricane Frances (2004). Data from the SMART-radars coupled with tower measurements have shed some light on the turbulent structure of hurricane winds and coherent features of alternating strong and weak wind speeds known as streaks.

The prolific Atlantic hurricane season's of 2004 and 2005 kept TTUHRT busy with seven deployments over the 2 seasons. In 2005, TTUHRT collected the only complete wind data record from Hurricane Katrina's Mississippi landfall. These seasons also highlighted the need for even more measurements in order to understand the variability in a hurricane's wind field at landfall. In 2005, the Atmospheric Science Group and the Wind Science and Engineering Research Center set out to develop a multi-use rapidly deployable surface observing station. A design competition was held within a graduate course in field measurements. What resulted was affectionately dubbed "StickNet" which is a 2.25 m rapidly deployable weather station which can be used to measure a variety of phenomena. StickNet platforms are far less rugged than the original WEMITE towers but are estimated to withstand a 3-second gust wind speed of 140 mph. The costs to produce a single StickNet probe are far less than a single WEMITE tower. TTUHRT currently maintains 24 StickNet probes. Although the probes are not the standard measuring height of 10 m, the group was willing to sacrifice this in order to gain a significantly larger number of observations from the landfall area. Additionally, the small footprint of StickNet probes allows them the versatility to be deployed in locations where the much larger WEMITE towers could not go. This allows for vital data collection closer to the immediate shoreline. The hurricane configuration of a StickNet allows for a deployment time of only 5-10 minutes while a WEMITE tower would take four people at least 1 hour. A good example of this came during Hurricane Dolly (2008), where a probe was deployed on a dune structure on South Padre Island. Hurricane Ike (2008) also provided a unique deployment opportunity at Fort Travis on the western tip of the Bolivar Peninsula. These types of deployments highlighted the benefits of the StickNet platforms. StickNet probes were first deployed into the hurricane environment for Hurricane Dolly (2008). Two other deployments were made in 2008 for Hurricane Gustav and Hurricane Ike. With the implementation of the StickNet project, the original WEMITE towers have been retired from service but remain in working order housed in the Wind Science and Engineering Research Center's Reese facility.

Excerpt from the National Hurricane Center's official cyclone report for Hurricane Ike...

"A 1-minute sustained wind of 71 kt was recorded by a Texas Tech University Hurricane Research Team (TTUHRT) anemometer near Winnie, Texas, between Houston and Beaumont. A 3-second gust of 95 kt was reported by a separate TTUHRT sensor near Hankamer, Texas. The pressure of 950 mb at landfall is based on a minimum pressure of 950.7 mb reported by a TTUHRT sensor at Port Bolivar near the entrance to Galveston Bay and a pressure of 951.7 mb reported at the Galveston Pleasure Pier"

In 2006 the Atmospheric Science Group and the Wind Science and Engineering were awarded a grant from the TTU Office of the Vice President for Research to develop two mobile Doppler radar systems for use in making measurements of boundary layer wind flows in a variety of conditions. TTUKA-1 was completed in time for the 2009 Atlantic hurricane season and made its first tropical deployment for Tropical Storm Ida. TTUKA-2 recently was completed and participated in the VORTEX 2 tornado field research project. The two radars make measurements at a finer scale than the SMART-radars allowing for a more detailed view of the structure of hurricane winds.

Following the 2008 hurricane season, scientists at institutions around the country who study hurricanes at landfall met to discuss future collaborations. What came out of this meeting was the Digital Hurricane Consortium (DHC). The consortium is under the oversight of the Applied Technology Council who works closely with the Federal Government. The goal of the consortium is to provide comprehensive data and analyses from landfalling hurricanes in order to mitigate the effects of these systems on life and property. The DHC includes: Texas Tech University, the Florida Coastal Monitoring Program, The Center for Severe Weather Research, University of Alabama at Huntsville, Oklahoma University, Louisiana State University, South Alabama University, University of Colorado, University of North Florida, and Notre Dame. Government partners include the USGS and NOAA's Hurricane Research Division.

Since the project's inception in 1998, data collected from TTUHRT platforms has directly appeared in 12 peer-reviewed journal publications, NHC and FEMA post-storm reports, 9 technical reports, and over 40 conference presentations and proceedings.

Well this was our story and how we came to be... in some later blogs we hope to talk about each of our instrument platforms in more detail as well as deployment strategies. We also hope to provide information regarding active cyclones in which landfall operations are being considered and provide updates from the field during deployments. The principal investigator and director of the project is Dr. John Schroeder. He is also the Director of the Wind Science and Engineering Research Center.